Discharge planning, self-management, and community support: Strategies to avoid psychiatric rehospitalisation from a service user perspective Patient Education and

Objective: Psychiatric rehospitalisation is often seen as a negative outcome in terms of healthcare quality and cost, as well as potentially hindering the process of recovery. The purpose of our study was to explore psychiatric rehospitalisation from a service-user perspective, paying attention to how rehospitalisation can be avoided. Method: Eight focus groups, including a total of 55 mental health service users, were conducted in six European countries (Austria, Finland, Italy, Norway, Romania, and Slovenia). The results were analysed using systematic text condensation. Results: All participants had been in touch with mental health services for at least one year, and had experienced more than one psychiatric hospitalisation. Participants emphasised the importance of discharge planning and psychoeducation both during and after the hospital stay, as well as the bene ﬁ ts of structured plans, coping strategies, self-monitoring techniques, and close contact with local community services.Social contacts and meaningful activities were also considered to be critical, as was support from peers and family members. Conclusion: Efforts to avoid psychiatric rehospitalisation should include actions that support a functional day-to-day life, improve coping strategies, and build on cross-sectoral collaboration. Practice implications: The study emphasises the need for psychoeducational and psychosocial interventions, starting already during the inpatient stay.


Introduction
Psychiatric rehospitalisation, sometimes called the 'revolving door' phenomenon, denotes a pattern of hospital admission in which individuals undergo a period of inpatient care and are then discharged back into the community, only to be rehospitalised for similar reasons within a relatively short period of time [1,2]. Reducing the reliance on inpatient care has been an important cornerstone of the deinstitutionalisation movement [3], with frequent psychiatric rehospitalisation often being related to poor service quality [4]. Psychiatric rehospitalisation is, however, likely to be more complex than this and attributable to many elements such as individual, system, and structural factors [5][6][7]. The phenomenon remains poorly understood, regardless of its negative implications on both cost and recovery [5,6]. A clearer understanding of psychiatric rehospitalisation is needed to achieve a balanced model for mental health services, including both community and hospital-based options [8]. Although research indicates higher levels of life satisfaction in the community setting [9] and a preference for patient-centred approaches [10,11], it is important to acknowledge that hospital admission may at times be necessary for recovery.
Scholars such as Durbin et al. have highlighted the need for more intervention studies on psychiatric rehospitalisation and the need for increased attention to discharge practices and community care [12]. The discharge process from psychiatric inpatient care has been described as stressful and re-adaptation into the community as overwhelming; both increase the risk of subsequent rehospitalisation [13][14][15][16][17]. Participants in a recent qualitative study depicted psychiatric rehospitalisation as the result of insufficient internal, interpersonal, and/or environmental resources [18], while other qualitative studies have reported psychiatric rehospitalisation as being beneficial, a necessity, or a relief, but also as lastresort preferences in cases where progress or recovery have stalled or regressed [18][19][20][21].
Including the perspectives of mental health service users has been acknowledged as essential to improving healthcare outcomes [22][23][24], although studies in relation to service users' views on psychiatric rehospitalisation remain few and far between [1,2]. To fill this gap, the current study explores service users' perspectives on psychiatric rehospitalisation, with particular attention to strategies that may decrease the need for its occurrence.

The CEPHOS-LINK project
The current paper is an extension of the EU-FP7 project known as Comparative Effectiveness Research on Psychiatric Hospitalisation by Record Linkage, or CEPHOS-LINK, which studied psychiatric rehospitalisation in six European countries, namely Austria, Finland, Italy, Norway, Romania, and Slovenia. Alongside a large register-based study, the CEPHOS-LINK project included a phenomenological study using focus groups to delineate mental health service users' views of psychiatric rehospitalisation [21]. A previous paper by Ådnanes et al. [21] explored the personal experiences of psychiatric rehospitalisation, and the current study continues this line of enquiry by exploring how psychiatric rehospitalisation can be avoided. The current study makes use of the same data set, using transcriptions from a total of eight focus groups comprising 55 participants. Three main topics of interest were included in the focus groups, two of which were explored in the previous paper by Ådnanes et al. [21]. The third topic, how to avoid rehospitalisation, is focus of the current analysis.

Sample
Participants were recruited through convenience sampling via mental health organisations and/or activity and day care centres in each country. Participants who had been in receipt of mental health services for at least one year and had experienced more than one psychiatric hospitalisation were invited to take part. The recruitment process was predominantly done by a dedicated contact person working in the user organisation or the activity or day care centre. Information and invitations were distributed via newsletter or announcements on home pages, via telephone, or face to face. In some cases, a researcher visited the organisation to provide information about the project, following up with invitations both verbally and in writing. In this way, a large number of potential participants received invitations to take part in the study, although it is not possible to gauge a response rate in this context.
In Romania and Austria, participants were recruited from the capital cities. In Finland and Slovenia, focus groups were held in the capital city, but included participants from all over the country. In Norway, the focus group was conducted in the third-largest city, and in Italy, focus groups took place in a north-eastern town.

Data collection
Focus groups were conducted between September 2016 and March 2017 either at the premises of a mental health organisation or at an activity or day care centre. Focus groups were led by a project researcher with expertise in mental health, working with assistant moderators who were either researchers and/or professionals from the mental health organisation. The focus groups lasted between 60 and 90 min. Moderators ensured that a good atmosphere was maintained during the interviews and that all participants contributed to the discussions with prompts and turn  Data saturation was considered to occur after eight focus groups when the themes became consistent, indicating a high probability that all the themes had been explored.

Data analysis
The results from the focus groups were transcribed verbatim, imported into the HyperRESEARCH 3.7.3 qualitative software, and analysed on principles of systematic text condensation [26]. In contrast to the first study [21], the present study focused on the third theme in the interview guide, which concerned 'how to avoid rehospitalisation'. New codes around this theme were generated and subsequently clustered to form descriptive themes. Descriptive themes considered to be related were further clustered to form analytical themes. All themes were grounded in the text throughout the analysis.
Validity in qualitative research is based on the 'appropriateness' of the tools, processes, and data [27]. Transcripts from all six countries were read by a smaller group of researchers in the team and agreement on all themes was reached via meetings to discuss data and analytic approaches. Transcripts were also read by several focus-group participants in Austria and Finland. Additionally, a representative from a Norwegian service-user organisation was actively involved in the entire research process, evaluating the interview guide, participating as a co-moderator, and discussing the results with researchers, in addition to co-authoring this article.

Results
A total of 55 participants took part in the current study, 60 % of whom were female. The participants' ages ranged from 26 to 65 years, and almost half reported living alone. Participants had experience of more than one psychiatric hospitalisation and had been in contact with mental health services for at least one year at the time of the focus groups. In terms of psychiatric diagnoses, 41.8 % reported psychotic disorders, 38.2 % bipolar disorder, 21.8 % depressive disorder, and 12.7 % anxiety disorder. Table 1 describes their main characteristics, which were rather consistent across countries.
The focus group findings were categorised into five themes that represented strategies related to avoiding rehospitalisation: (1) plans and preparation for discharge; (2) self-monitoring and individual coping; (3) follow-up by outpatient services and other community services; (4) being part of social community and doing meaningful activities; and (5) informal support, including peer support and family support (in addition to support from mental health services). Table 2 illustrates the five themes and subthemes, including examples of quotes extracted from the focus group transcription.

Plans and preparations for discharge
Many participants found coping in the community after an inpatient stay to be challenging. Several participants highlighted the importance of planning and preparing for hospital discharge during the hospital stay. Participants mentioned that this was rarely done, adding that this lack of adequate planning increased the risk of crises and rehospitalisations. Some participants emphasised the importance of going through what services were available in the community and what to expect in terms of followup appointments. Some participants described beneficial discussions during their inpatient stay with healthcare professionals who outlined treatment options and provided advice on how to develop a supportive social network. Several participants mentioned the need for meaningfulness in day-to-day life as well the importance of preventing loneliness and lack of structure. Participants also mentioned the importance of individual planning such as developing emergency or crises plans prior to discharge.

Self-monitoring and individual coping
Several participants stressed the significance of having coping strategies and the need for self-monitoring to identify signs of deteriorating mental health. Such crisis planning outlined what to do if symptoms worsened, with participants feeling an enormous sense of safety when they had someone to call. Austrian participants also mentioned specific skills training as a useful tool for planning and developing coping strategies.
Many participants admitted to a lack of understanding with respect to medications and expressed the need for more information in this context-for example, how to take medications as prescribed and how dosages and combinations of medications can be adjusted. This balance was considered by several participants to be an important factor in preventing rehospitalisation.
Some participants also talked about the importance of things they could do in their daily lives to avoid rehospitalisation, such as taking good care of themselves both physically and mentally. They mentioned as being useful such simple things as eating healthy, exercising, going for a walk, and having regular social contacts, and noted that this is reflected in most literature on mental health promotion.

Follow-up by outpatient and other community services
The participants agreed on the need for sufficient follow-up after hospital discharge. Most participants reported receiving some level of follow-up, via outpatient appointments with such healthcare professionals as psychiatrists, psychologists, or general practitioners (GPs). These appointments were described as important for treating symptoms and controlling medication, although several participants felt that these visits were not frequent or in-depth enough, describing these visits as 'just talking, and then it is back to reality', with others being quite critical of what they perceived as an overuse of medication. Several participants mentioned the lack of available professionals and services and called for a more comprehensive approach, with closer monitoring of well-being to identify changes or developments in their behaviour or condition. Austrian participants brought up an example of good practice in the form of a 'psychosocial service' that prevented rehospitalisations by following up with individuals three to four times a week. Italian participants, on the other hand, mentioned the benefits of day care centres, which also afforded them access to mental health professionals. Romanian participants endorsed home visits provided by NGOs as useful if the doctors were busy. Other follow-up actions such as telephone calls each morning were found to be helpful for structuring daily life; other participants mentioned feeling 'safer' simply by having a telephone number available to call in case of a crisis, even if they never used it. Austrian participants highlighted the need for an 'intermediate service' between mental health services and socially focused services, something that would provide support for daily life in terms of outreach services and home visits.

Being part of social community and doing meaningful activities
Many participants expressed the desire to live on their own in their homes, describing the benefits of being part of a community and having access to meaningful activities for avoiding psychiatric rehospitalisation. Having 'something to do' after being discharged from hospital was mentioned in all the focus group interviews, although participants used different terminology to describe these activities, depending on the country. Most described a combination of structured activities (such as work) and social opportunities as provided by private organisations, NGOs, or public services; these provided structure, activity, training of different kinds, and social contacts, and added meaningfulness to service users' daily life. Some of these even allowed overnight stays-'crisis beds'-to forestall the need for full-scale rehospitalisation.
Different types of residential services, such as long-term residential care and assisted living arrangements, were mentioned by a few participants as important factors in avoiding rehospitalisation.

Peer and family support alongside professional help
Another useful way of preventing psychiatric rehospitalisation was through user-based initiatives, including peer support. Several participants had experiences with these kinds of initiatives; interestingly, they reported that peer support felt 'safer' than professional support because participation in support groups did not lead to enforced hospital admission. Participants discussed different models of peer support and peer contact such as those that had started during their hospital stays. Some participants also talked about the need for combining peer support and professional help, and they said that these should be seen as complementary.
Support from family was also mentioned by many, with some participants describing family as very important and others describing experiences of abandonment or more or less breaking contact with close family members.

Discussion
The participants articulated several ways of avoiding psychiatric rehospitalisation on many different levels. They talked about the importance of adequate preparation through obtaining information and making plans prior to discharge from hospital. Discharge planning is a strategy that that has been widely endorsed in different contexts [28], although its effect on rehospitalisation rates, lengths hospital of stay, health outcomes, and costs remains uncertain [29][30][31]. Discharge planning has also been found to ease transition [28] by preventing feelings of isolation. Participants in the current study echoed results from a recent review by Vigod et al. [32], who found that pre-and postdischarge psychoeducation, structured needs assessments, medication reconciliation and education, transition managers, and inpatient/outpatient provider communication were useful components in a successful transition from hospital to community. The aim of psychoeducational interventions is to improve outcomes by helping service users to develop knowledge and understanding of symptoms, treatment options, and coping strategies. A Cochrane report on psychoeducation for schizophrenia concluded that psychoeducational interventions seemed to reduce relapse and readmission, reduce the length of hospital stays, and encourage medication compliance [33].
A review by Nurjannah found the most important factor in effective discharge planning to be communication between the different stakeholders within the mental health system, but notably among family members [30]. The review highlighted the need for further research on how communication during discharge planning might benefit future health outcomes. Shared decision-making (SDM) facilitates patient engagement in treatment decision-making, and this was explored in a recent pilot study that used SDM for choosing community psychiatric rehabilitation services prior to discharge. Although the study did not find an effect in terms of rehospitalisation rates [34], it did find that SDM led to improvements in personal recovery and concluded that SDM should be an integral part of the discharge process.
A recent review outlined the importance of autonomy and the benefits of empowering patients to feel less dependent on the mental health system while still receiving adequate support [17]. Coping strategies may depend on individual contexts; for example, Rise et al. found coping strategies to be more useful during periods of fewer symptoms, with flexible, safe, and predictable support options being more important during periods when symptoms were more severe [20].
Participants in the current study mentioned the need to develop knowledge about medications and coping strategies, both of which have been the subject of previous studies [35,36]. However, relatively little is known about psychoeducation in inpatient settings, and more research is needed, particularly in terms of how best to target specific patients at different stages of treatment [36]. New models promoting autonomy and empowerment have produced interesting results, Moljord et al. [37] reported that self-referral to inpatient treatment did not have a significant effect on the number of inpatient days or the number of hospitalisations; however, self-referral appeared to encourage patients' engagement and participation. This finding supports a fundamental shift in provider-service user relationships, with a stronger emphasis on encouraging service users' engagement [38].
Participants in the current study emphasised the need for more extensive and more frequent follow-up from outpatient or community mental health services. Medication was identified as an important factor in avoiding rehospitalisation, and this was also mentioned in a recent UK study [39]. Different formats of outpatient contacts were cited as reducing the risk of rehospitalisation by providing a sense of safety in day-to-day life and the feeling that their well-being was being monitored appropriately. Previous studies, such as a review by Sfetcu et al. [7], showed that service users having direct contact with a primary care provider reduced psychiatric rehospitalisation, although their just being registered with a GP had no impact. The same review also found a positive association between the risk of rehospitalisation and the intensity of outpatient contacts [7].
The importance of meaningful social activities, such as those offered by day or activity centres, was articulated by the participants in the current study. This finding concurs with previous studies reporting positive impacts from communitybased social activities on the quality of life [40,41]. Silva et al. [42] reported a 20 % lower risk of rehospitalisation for patients referred to community-based psychosocial support units following inpatient care, in comparison with patients referred to the usual formats of outpatient care. Silva et al. also called for further research to identify which specific components of psychosocial services led to a reduction in rehospitalisations. Participants in the current study reported that different types of innovative community-based services (e.g. public, private and NGO) were particularly useful, lining up with previous studies [19,37] that highlighted the benefits of actions that empower service users, provide new opportunities, and build bridges into the wider community [43].
Most participants in the current study mentioned the effectiveness of support from peers and from family in avoiding rehospitalisation. In previous studies, Sledge et al. found that patients assigned peer mentors had significantly fewer rehospitalisations than those without [44,45]. This was echoed by Johnson et al. [46], who reported that self-management interventions by peer support workers reduced the number of acute psychiatric rehospitalisations. Peer workers have been found to be useful coeducators, improving patient activation and treatment participation [47,48]. Although the supportive nature of family was highlighted by the participants in the present study, this was limited to positive family relationships. Family as a protective mechanism in terms of relapse and rehospitalisation has been well-documented, [7,49]; however, maladaptive family relationships, such as having family members who hold negative attitudes toward mental health service users or who view mental health issues as a stigma, have been found to increase these risk relapse and rehospitalisation [7,50].
As in a previous article [21] from the CEPHOS-LINK project, the current study had both strengths and limitations. One area of limitation relates to sampling. Although it was crucial to include participants who were happy to discuss their experiences, relying on convenience sampling could have resulted in a participant group representing a higher-than-average level of social functioning or presenting a particular narrative. The authors were conscious of the potential selection bias that may have arisen from the recruitment process. One balancing factor for this was the heterogeneity of the participant group, which represented different sociodemographic and clinical groups. Another limitation of the study was that although the inclusion criteria specified that participants should have experienced at least one rehospitalisation, it did not specify the amount of time between hospitalisations; therefore, it was not clear whether the rehospitalisations occurred within a short or long period of time. It is also noteworthy that the participants themselves did not discuss the amounts of time between rehospitalisations. This should be explored further.
A clear strength of this study was that it included experiences from mental health service users from multiple European countries with different healthcare systems. Although a detailed comparison across countries was not possible due to the limited number of participants from each country, it is striking that the experiences of participants from all six countries appeared to be remarkably consistent.
The validity of the data used in this study was safeguarded in several ways. First, the study design was identical for all six countries; it was defined by an interview guide and performed by experienced researchers. Second, the resulting transcriptions were read by several of the focus group participants, as well as by a service-user representative who participated in the entire research process. Finally, the analyses were performed by incorporating interpretations from different perspectives.

Conclusion
This qualitative study indicated that steps to avoid psychiatric rehospitalisation are possible and can be taken on many different levels. Most of the focus group discussions concurred that while services users want to live at home, they may need considerable support to cope with their mental health difficulties. Preparing for life after discharge during their hospital stays was mentioned as particularly useful for avoiding psychiatric rehospitalisation, as was improving professionals and community outpatient follow-up and access to innovative models of coping strategies for selfmanagement. The findings of the current study emphasised the need for meaningful activities in daily life and the benefits of including peers and family as sources of support in collaboration with healthcare professionals. Striking similarities in service users' experiences across all six countries suggested that these needs and insights are relevant irrespective of the healthcare system or other contextual aspects.

Practice implications
Our results have provided us with a deeper understanding of the internal, interpersonal, and community aspects of psychiatric rehospitalisation. Results from this study may inform the development of discharge planning practices-for example, by encouraging shared decision making, self-management training, and coping strategies, as well as by highlighting the importance of peer support and family actions, all of which should start during the hospital stay. Discharge plans should be tailored to the patients' needs and should include psychoeducative actions detailing signs of crisis and possible relapse and outline appropriate responses. Strategies for self-monitoring and coping should be implemented both before and after discharge, with support from multisector professionals such as mental health providers and social workers.
In the community, close follow-up is crucial, and discharge plans should include tailored follow-up visits shortly after discharge, according to the patients' needs.
A multisector approach can be useful in supporting continuity of care and should make use of both outpatient and psychosocial resources, including NGO actions to empower service users and their families with new opportunities and bridges into the community. Further studies are needed to identify the specific components of psychosocial services that support recovery and reduce rehospitalisation.
Findings from the current as well as previous studies reinforce the interplay between mental health, interpersonal relationships, and social determinants of mental health [51]. These factors point to potential initiatives that can support people in their own communities. National policies should reflect these approaches and advocate for further development of multisector actions [52].
Qualitative approaches can shed important light on the subjective experience of complex phenomena such as psychiatric rehospitalisations by incorporating knowledge from lived experiences, opening up new avenues for more person-centred approaches in both the community and the hospital context.

Availability of data and material
To preserve participant anonymity, the dataset generated and analysed during the current study will not be shared.

Declaration of Competing Interest
The authors declare that they have no conflicts of interest.

Funding
This study was funded by the European Union's Seventh Framework Programme for Health under grant agreement number 603264.

Credit author statement
MÅ designed the focus group study. MÅ, JCS, VD, HW, LS, RS and CS collected the data. MÅ and LM did the first analyses of the data. MÅ, JCS and VD followed up with more thorough analysis and interpretation. MÅ, VD and JCS wrote the manuscript. All authors contributed in revising the manuscript for important intellectual content. JCS edited the text. All authors read and approved the final manuscript.